Review

Challenges Associated with Implementing 5Gin Manufacturing

Eoin O’Connell 1,2,* , Denis Moore 1 and Thomas Newe 1,2,*1 Electronic & Computer Engineering Department, University of Limerick, V94 T9PX Limerick, Ireland;

denis.moore@ul.ie2 CONFIRM Centre for Smart Manufacturing, University of Limerick, V94 T9PX Limerick, Ireland* Correspondence: eoin.oconnell@ul.ie (E.O.); thomas.newe@ul.ie (T.N.)

Received: 24 April 2020; Accepted: 9 June 2020; Published: 12 June 2020�����������������

Abstract: 5G networks will change several industries, including manufacturing. 5G has the potentialto become the future communication platform of choice for many industries and in particular themanufacturing sector, driving the future of Industry 4.0 and smart manufacturing. The vision of a“factory of the future” is now tangible for many industry sectors. The ability to cope with increasedbandwidth, latency requirements, big-data generated from more connected equipment and the dataprocessing required on the factory floor is a massive challenge for industry. This paper discusses how5G can impact a manufacturing environment, the standards and technical requirements needed tomeet the demands of utilizing 5G, and the security issues that need to be addressed if planning a5G deployment.

Keywords: Internet of Things; 5G; spectrum; manufacturing; private network; edge; core; MPN;5G security; industry 4.0; smart manufacturing

1. Introduction

5G promises to deliver faster download speeds, lower latency and higher capacity, as shown inTable 1; Table 2 below. The goal of Industry 4.0 is to maximize efficiency by utilizing these featuresacross all processes and assets at all times; to provide a better understanding of the manufacturingprocesses across their manufacturing sites in near real-time. With the expected increase in datarequirements ranging from mission-critical to massive machine connectivity, the deployment of 5G hasraised expectations that it will open new opportunities for manufacturing business models.

Table 1. Data Speeds for 3G, 4G and 5G.

Network Type Average Download Speeds Peak Download Speeds Theoretical Download Speeds

3G 8 Mbps ~20 Mbps 42 Mbps

4G 32.5 Mbps 90+ Mbps 300 Mbps

5G 130Mbps–240 Mbps 599 Mbps+ 10–50 Gbps

Table 2. Comparison of 3G/4G and 5G latency times [1].

Network Type Milliseconds (ms)

3G Network 60 ms (Typical)

4G Network 50 ms (Typical)

5G Network 1 ms (theoretical)

Telecom 2020, 1, 48–67; doi:10.3390/telecom1010005 www.mdpi.com/journal/telecom

http://www.mdpi.com/journal/telecomhttp://www.mdpi.comhttps://orcid.org/0000-0002-3173-140Xhttps://orcid.org/0000-0002-3375-8200http://dx.doi.org/10.3390/telecom1010005http://www.mdpi.com/journal/telecomhttps://www.mdpi.com/2673-4001/1/1/5?type=check_update&version=2

Telecom 2020, 1 49

According to McCann et al. [2], “The Fourth Industrial Revolution or simply ‘Industry 4.0′ ishow manufacturing industry expects to maximise the innovations of 5G wireless communicationsby automating industrial technologies and utilising other enabling technologies such as artificialintelligence (AI) and machine learning. Industry expects this to lead to more accurate decision makingsuch as automation of physical tasks based on historical information and knowledge, or improvedoutcomes for a wide range of vertical marketplaces not just in manufacturing but verticals such asagriculture, supply chain logistics, healthcare, energy management and an ever-increasing number ofindustries becoming more aware of the potentials of 5G.” It is expected to achieve this in three waysthrough the use of network slicing, Figure 1.

• eMBB (Enhanced Mobile Broadband)• URLLC (Ultra-Reliable Low Latency Communications)• mMTC (Massive Machine-Type Communications)

Telecom 2020, 1,  2 

According to McCann et al. [2], “The Fourth Industrial Revolution or simply  ‘Industry 4.0′  is how manufacturing industry expects to maximise the innovations of 5G wireless communications by automating  industrial  technologies  and  utilising  other  enabling  technologies  such  as  artificial intelligence  (AI)  and machine  learning.  Industry  expects  this  to  lead  to more  accurate  decision making  such as automation of physical  tasks based on historical  information and knowledge, or improved outcomes for a wide range of vertical marketplaces not just in manufacturing but verticals such as agriculture, supply chain logistics, healthcare, energy management and an ever‐increasing number of industries becoming more aware of the potentials of 5G.” It is expected to achieve this in three ways through the use of network slicing, Figure 1. 

eMBB (Enhanced Mobile Broadband)  URLLC (Ultra‐Reliable Low Latency Communications)  mMTC (Massive Machine‐Type Communications) 

 Figure 1. 5G key usage scenarios. 

1.1. Enhanced Mobile Broadband (eMBB)   

eMBB can initially be addressed by the expansion to existing 4G services as it aims to service more densely populated metropolitan centers with downlink speeds approaching 1 Gbps (gigabits‐per‐second) indoors, and 300 Mbps (megabits‐per‐second) outdoors. One way to accomplish this is through  the  installation  of  extremely  high‐frequency millimeter‐wave  (mm‐Wave)  antennas.  For areas  that are more urban outlying and beyond  into rural setting areas, eMBB will work  towards replacing 4G’s  current LTE  (Long‐Term Evolution)  scheme, with a new network of  lower‐power omnidirectional antennas providing 50 Mbps downlink service. eMBB traffic can be deemed to be an addition to the 4G broadband service currently available. It is capable of large payloads and a by‐a‐device activation pattern  that  is capable of  remaining  stable over an extended  time  interval. This allows  the network  to schedule wireless resources  to  the eMBB enabled devices such  that no  two eMBB  devices  compete  or  access  the  same  resource  simultaneously  [3].  To  provide  these,  it  is expected that 5G will deliver: 

Traffic capacity of up to 10 Mbps per sq. metre in hotspot areas.  Data transfer rates the user experiences of up to 1 Gbps with peak data rates of transfer in the 

tens of Gbps and a capacity of maximum traffic volume of at least 1 Tbps per sq. kilometre.  Latency as low as 1ms for user experience levels of data exchange.  Connection density of up to one million connections per sq. kilometre.  High mobility, facilitating connectivity up to 500 km/h in high‐speed trains and up to 1000 km/h 

in aeroplanes, with enhanced user experience. 

1.2. Ultra‐Reliable and Low‐Latency Communications (URLLC)   

URLLC can address critical needs of communications where bandwidth is not quite as crucial as speed, e.g., an end‐to‐end latency of 1 ms or less. The design of a low‐latency and high‐reliability 

Figure 1. 5G key usage scenarios.

1.1. Enhanced Mobile Broadband (eMBB)

eMBB can initially be addressed by the expansion to existing 4G services as it aims toservice more densely populated metropolitan centers with downlink speeds approaching 1 Gbps(gigabits-per-second) indoors, and 300 Mbps (megabits-per-second) outdoors. One way to accomplishthis is through the installation of extremely high-frequency millimeter-wave (mm-Wave) antennas.For areas that are more urban outlying and beyond into rural setting areas, eMBB will work towardsreplacing 4G’s current LTE (Long-Term Evolution) scheme, with a new network of lower-poweromnidirectional antennas providing 50 Mbps downlink service. eMBB traffic can be deemed to bean addition to the 4G broadband service currently available. It is capable of large payloads and aby-a-device activation pattern that is capable of remaining stable over an extended time interval.This allows the network to schedule wireless resources to the eMBB enabled devices such that no twoeMBB devices compete or access the same resource simultaneously [3]. To provide these, it is expectedthat 5G will deliver:

• Traffic capacity of up to 10 Mbps per sq. metre in hotspot areas.• Data transfer rates the user experiences of up to 1 Gbps with peak data rates of transfer in the tens

of Gbps and a capacity of maximum traffic volume of at least 1 Tbps per sq. kilometre.• Latency as low as 1ms for user experience levels of data exchange.• Connection density of up to one million connections per sq. kilometre.• High mobility, facilitating connectivity up to 500 km/h in high-speed trains and up to 1000 km/h

in aeroplanes, with enhanced user experience.

Telecom 2020, 1 50

1.2. Ultra-Reliable and Low-Latency Communications (URLLC)

URLLC can address critical needs of communications where bandwidth is not quite as crucialas speed, e.g., an end-to-end latency of 1 ms or less. The design of a low-latency and high-reliabilityservice involves several components, including, incredibly fast data turnaround, efficient control anddata resource sharing, grant-free based uplink transmission, and advanced channel coding schemes.Uplink grant-free structures guarantee a reduction in user equipment (UE) latency transmissionby avoiding the middle-man process of acquiring a dedicated scheduling grant. These servicesare supported by the 5G New Radio (NR) standard. URLLC should be the tier that addresses theautonomous vehicle category, where decision time for a reaction to a possible accident needs tobe almost non-existent. URLLC makes 5G a possible competing solution with satellite, for GlobalPositioning System (GPS) geolocation services [4,5].

1.3. Massive Machine-Type Communications (mMTC)

5G offers massive machine-type communication (mMTC), which aims to support tens of billions ofnetwork-enabled devices to be wirelessly connected [1]. Today’s communication systems already servemany MTC applications. However, the characteristic properties of mMTC, i.e., the massive number ofdevices and the tiny payload sizes, require novel approaches and concepts. 5G allows a density ofone million devices per square kilometer [3,6]. 5G will be able to carry a lot more data and transferit much faster than 4G LTE, however, faster is not always better or even necessary in the Internetof Things (IoT) world, especially when it typically requires more power on the end device. So the5G NR standard will introduce new device types, like Cat-M1 (operates at 1.4 MHz bandwidth) andnarrow band (NB-IoT). Both NB-IoT and Cat-M1 devices can sleep for extended periods of time withextended discontinuous reception (eDRX) and power-saving mode (PSM) functionalities, which greatlyreduce device power consumption This will enable lower-power, and battery-driven devices to beutilized. 5G in industrial automation will facilitate real-time wireless sensor networks with locationand asset tracking. 5G networks will eventually be able to supersede wired connections in even themost demanding of applications, such as automation control and high throughput vision systems [7,8].

This paper differs from previous publications in this space in that it provides an overview of 5G andthe technology required, how it will positively impact the future of the manufacturing sector, enabling itto move to industry 4.0 and the infrastructure requirement to enable an industry to adopt 5G. The paperis structured as follows: Section 2 discusses the potential and impact of 5G in manufacturing; Section 3presents the use cases of 5G in manufacturing while Section 4 lists the infrastructure requirements fora 5G network; Section 5 discusses where 5G lies in the wireless spectrum and Section 6 outlines theconcept and benefits of private 5G networks; Section 7 briefly discuss 5G security issues and Section 8the 5G standard; Finally our conclusions regarding 5G and its role in manufacturing are presented.

2. The Potential of 5G in Manufacturing

5G has the potential to become the underlying digital fabric connecting all elements of themanufacturing world. 5G promises fast connectivity, more bandwidth than Wi-Fi and 4G LTE butalso low latency with support for thousands of devices in a small location, all of which are attractiveprospects to manufacturing facilities. 5G’s integration capability with time-sensitive networking (TSN)makes smart factories an achievable goal. TSN capability will help 5G avail of the enhancements to theethernet to make it more reactive to demand or “deterministic”, essentially making 5G available ontime with low latency, low jitter and high reliability. In order for this interface to achieve low latency,all the devices have to synchronize to the same time-base, in effect they have to be ‘time aware’.

Widespread rollout of 5G is now a reality. Device manufacturers have already started deploying5G network equipment while network carrier companies are on the verge of commercializing 5Gnetworks in new verticals such as manufacturing [9]. While there are several barriers still impedingthe universal deployment of 5G networks, there are also endless applications and potential benefits.

Telecom 2020, 1 51

5G has the potential to revolutionize various critical areas such as the IoT, entertainment, automation,IT, and more. 5G private networks, which offers several advantages, will boost the power of IndustrialIoT (IIoT) as 5G offers more than increased network speeds and bandwidth [10].

5G wireless technology and its associated deployment is being implemented at a time when manyindustries are themselves transforming through greater automation, using techniques such as the IoT,machine vision, machine learning and robotics. 5G connections will go beyond human communicationneeds and is expected to enable the intelligent internet of things. The next generation (NextGen) oftelecommunication technologies will be adopted by a broader range of industries and sectors withdigital manufacturing a big beneficiary. NextGen wireless control of highly sophisticated mobilemachines is becoming feasible, and it will allow the possibility for machines to take advantage of theenormous computing power and cloud storage options available, without being tethered by physicalwires. An early area of interest for 5G in the IIOT is to facilitate factory monitoring, where sensors areinstalled in the factory to monitor equipment conditions and to oversee that everything is working asit should, and to identify potential issues before they occur. These sensors will be interrogated andthe end data used to create insights into end products and deliverables, providing improvementsin quality control and bespoke manufacturing. 5G has incorporated many requests to improve 4Gand now is capable of delivering new services that may be tuned to the requirements of a customers’business models (Figure 2) e.g., manufacturing [7,8].

Telecom 2020, 1,  4 

5G wireless  technology and  its associated deployment  is being  implemented at a  time when many industries are themselves transforming through greater automation, using techniques such as the  IoT, machine  vision, machine  learning  and  robotics.  5G  connections will  go  beyond  human communication needs and is expected to enable the intelligent internet of things. The next generation (NextGen) of telecommunication technologies will be adopted by a broader range of industries and sectors with digital manufacturing a big beneficiary. NextGen wireless control of highly sophisticated mobile machines is becoming feasible, and it will allow the possibility for machines to take advantage of the enormous computing power and cloud storage options available, without being tethered by physical wires. An early area of interest for 5G in the IIOT is to facilitate factory monitoring, where sensors are installed in the factory to monitor equipment conditions and to oversee that everything is working as  it  should, and  to  identify potential  issues before  they occur. These  sensors will be interrogated and the end data used to create insights into end products and deliverables, providing improvements in quality control and bespoke manufacturing. 5G has incorporated many requests to improve 4G and now is capable of delivering new services that may be tuned to the requirements of a customers’ business models (Figure 2) e.g., manufacturing [7,8].   

 Figure 2. Use cases—latency and bandwidth requirements [8]. 

Data  rates of  10 Gbps  are widely  accepted  as  a  feasible  expectation  for  5G when  it  is  fully commercially available, but early 5G services seem to be maxing out at approximately 1 Gbps [11]. As deployments  increase,  all  the data generated will  be  collected  and  analyzed  to  give detailed insights into every facet of a factory’s operation. Machine learning‐capable smart robots will build products and be enabled to move goods and deploy equipment in and around the factory, generating even more data. Manufacturing is expected to pivot from an assembly line to the production of highly customized products essentially enabling the mass production of bespoke products, with workers streaming virtual/augmented reality videos or working in virtual environments [12]. These concepts can be facilitated by the 5G NR standard.   

While the range of benefits of 5G is impressive, there are limitations inherent in the deployment of 5G. The spectrum allocated is at a higher frequency, and proportionately the wavelength of those signals is smaller, which is limiting to the signal penetration capability. 5G will initially be allocated spectrum at the 3 to 6 GHz range (depending on national regulations). Within a heavy engineering environment, with dense, generally metallic machinery, signal penetration should be factored into the design. To calculate the penetration of a 5G signal the wavelength of the signal is needed, as signal 

Figure 2. Use cases—latency and bandwidth requirements [8].

Data rates of 10 Gbps are widely accepted as a feasible expectation for 5G when it is fullycommercially available, but early 5G services seem to be maxing out at approximately 1 Gbps [11].As deployments increase, all the data generated will be collected and analyzed to give detailed insightsinto every facet of a factory’s operation. Machine learning-capable smart robots will build productsand be enabled to move goods and deploy equipment in and around the factory, generating evenmore data. Manufacturing is expected to pivot from an assembly line to the production of highlycustomized products essentially enabling the mass production of bespoke products, with workersstreaming virtual/augmented reality videos or working in virtual environments [12]. These conceptscan be facilitated by the 5G NR standard.

While the range of benefits of 5G is impressive, there are limitations inherent in the deploymentof 5G. The spectrum allocated is at a higher frequency, and proportionately the wavelength of those

Telecom 2020, 1 52

signals is smaller, which is limiting to the signal penetration capability. 5G will initially be allocatedspectrum at the 3 to 6 GHz range (depending on national regulations). Within a heavy engineeringenvironment, with dense, generally metallic machinery, signal penetration should be factored into thedesign. To calculate the penetration of a 5G signal the wavelength of the signal is needed, as signalpenetration is proportional to its wavelength, i.e., longer wavelengths penetrate obstacles better. This iscalculated by using the frequency and the speed of light (in the medium), in Equation (1) [13].

λ = c /f (1)

where;

λ (Lambda) = Wavelength in metersc = Speed of Light (299,792,458 m/s)f = Frequency (MHz)

Applying the formula above, it becomes clear that there may be issues around signal penetrationand distance issues when transmitting at the bands allocated for 5G. As signal penetration is directlyrelated to signal wavelength, shorter wavelengths are more easily reflected or refracted. As thefrequencies utilized by 5G increases, the signal penetration decreases, i.e., it is more susceptible toattenuation. Accordingly, manufacturing should choose their operating band wisely as it may have asignificant impact. A summary comparison of 5G and 4G LTE (Long-Term Evolution) is shown belowin Figure 3.

Telecom 2020, 1,  5 

penetration is proportional to its wavelength, i.e., longer wavelengths penetrate obstacles better. This is calculated by using the frequency and the speed of light (in the medium), in Equation (1) [13]. 

λ = c /f  (1)

where; λ (Lambda) = Wavelength in meters c = Speed of Light (299,792,458 m/s) f = Frequency (MHz) 

Applying the formula above, it becomes clear that there may be issues around signal penetration and distance issues when transmitting at the bands allocated for 5G. As signal penetration is directly related  to  signal wavelength,  shorter wavelengths  are more  easily  reflected  or  refracted. As  the frequencies utilized by 5G increases, the signal penetration decreases, i.e., it is more susceptible to attenuation. Accordingly, manufacturing should choose their operating band wisely as it may have a significant  impact. A summary comparison of 5G and 4G LTE (Long‐Term Evolution)  is shown below in Figure 3. 

 Figure 3. Comparison of 4G LTE versus 5G. 

How 5G Will Impact Manufacturing 

For  5G  to  be  deployed  successfully  in  a manufacturing  environment,  there must  be  close collaboration between all systems, spanning across the corporate  information and communication technologies (IT) and the manufacturing operational technologies (OT) to maximize the capabilities inherent in 5G. The vision is that industry will be able to construct cyber‐physical systems that enable precision control in near real time, from virtually anywhere in the world [14]. Typical deployment scenarios, some of which were mentioned previously, are given below: 

Ability to remotely operate equipment e.g., production line robotics on the factory floor. Typical applications being welding, painting and assembly. 

Remote control of supply chain equipment e.g., operator can remote control equipment such as untethered  robots,  typical  applications  include,  autonomous  ground  vehicles  (AGVs)  or forklifts. 

Remote monitoring of equipment e.g., the transmission of diagnostics information, so service technicians arrive prepared for successful repairs and updates when needed. 

Figure 3. Comparison of 4G LTE versus 5G.

How 5G Will Impact Manufacturing

For 5G to be deployed successfully in a manufacturing environment, there must be closecollaboration between all systems, spanning across the corporate information and communicationtechnologies (IT) and the manufacturing operational technologies (OT) to maximize the capabilitiesinherent in 5G. The vision is that industry will be able to construct cyber-physical systems that enableprecision control in near real time, from virtually anywhere in the world [14]. Typical deploymentscenarios, some of which were mentioned previously, are given below:

Telecom 2020, 1 53

• Ability to remotely operate equipment e.g., production line robotics on the factory floor.Typical applications being welding, painting and assembly.

• Remote control of supply chain equipment e.g., operator can remote control equipment such asuntethered robots, typical applications include, autonomous ground vehicles (AGVs) or forklifts.

• Remote monitoring of equipment e.g., the transmission of diagnostics information, so servicetechnicians arrive prepared for successful repairs and updates when needed.

• Machine to machine communication: closed-loop communications between machines to optimizemanufacturing processes.

• Intra- and inter-enterprise communication: for monitoring of assets distributed in broadergeographical areas across the value chain.

• Augmented reality support in design, maintenance and repair: use of augmented reality to aidin the execution of procedural tasks in the design, maintenance and repair domain (throughsimulations).

Each industry deployment has to look at multiple constraints and identify a solution that willdeliver an optimal solution to meet the business requirements of the end-user or corporate need.

The main challenges for industry adoption of 5G are:

• Cost: the manufacturing industry has high-cost reduction requirements and will only implementnew applications if these have been proven to reduce costs ultimately.

• Safety: hundreds of connected automated devices on a factory floor can create a hazardousenvironment for humans.

• Deployment knowledge: many small and medium enterprises (SMEs) do not have the capacityto resource the learning requirements or the technical ability to capitalise on the potentials of 5G.

• RF interference: several objects on the factory floor are already using radio communications.

Another consideration for manufacturing is the increased proliferation of more sophisticatedcollaborative robots (cobots). These smaller, lightweight, smart robots can detect when a human isnearby, making it possible for that individual to work more safely alongside the machine. A typicalusage scenario for a cobot might be to take care of such tasks as moving heavy parts or doing repetitivetasks, while a human collaborator handles the more advanced tasks. Cobots are already on the market,but 5G networks should help make them more useful. An example of adding value is the capitalizingon the lower latency of 5G which will allow a more instantaneous real-time presence awareness thatwould enable the robot to work at faster speeds next to a human with a reduced risk (close to zero) ofinjury [15,16].

3. Use Cases of 5G in Manufacturing

While 3G and 4G offer stepped improvements in speed and bandwidth, 5G will be the first cellularplatform to provide reliable machine-to-machine and Industrial IoT systems. There are many potentialuse cases for 5G in Manufacturing.

3.1. Supply Chain Unification

Digitalizing the modern manufacturing unit, introducing a higher degree of automation controland the advent of smart Internet of Things devices is a focal point of the fourth industrial revolution.The introduction of 5G into the manufacturing industry enables supply chains to pivot from a dispersedseries of independently managed location networks to an increasingly connected ecosystem of devicesthat share knowledge across the network in pseudo real-time. As manufacturing units become morecapable with the integration of 5G into smart devices, the manufacturing industry can leverage theadvanced communication capabilities between devices, unifying the production process. 5G is rapidlybecoming the connection of choice, serving as the intermediary medium of data transfer between everycomponent of the manufacturing ecosystem [17].

Telecom 2020, 1 54

3.2. Artificial Intelligence (AI) and Fast Decision Making

The best-run processing plants depend on immense information pools to decide—with inescapabledeferrals as information is gathered, cleaned, and investigated. 5G accelerates the choice processduration, permitting enormous measures of data to be ingested, handled and actioned close toreal-time. In a few substantial enterprises, for instance, producers have had the option to selloverabundance vitality back to the matrix when machines are not running, and costs are reasonable.5G will permit the gushing of information progressively to the cloud, and the utilization of live videoexamination. For instance, a surveillance camera could see an unsettling influence, recognize if there isa fast-approaching risk or threat via AI, and dispatch an automaton or alarm a laborer to examine.

Modern-day manufacturing is based on multiple technologies working together, which translatesto the creation of large data sets being created. This large increase in data collection can be transmittedand analyzed faster with 5G’s low latency coupled with its high bandwidth capacity. The expansion ofmultiple sensors to machine monitoring by industrial facilities is creating more information than in thepast. Transmission through wired systems is costly compared to wireless networks. 5G can help byempowering the utilization of up-to-date information at scale [18]. While the production floor is theobvious point at which manufacturing can avail of the benefits of 5G there are also other examples ofhow 5G can contribute to the manufacturing environment in a non-direct manner. Examples of areaswithin an organization that could benefit indirectly are listed in Table 3.

Table 3. Use cases in manufacturing [2].

Use-Case Category Scenario Impact

Time-critical processes

• Real-time, closed-looprobotic control

• Video-drivenmachine-human interaction

• Augmented Reality/ VirtualReality (AR/VR) formaintenance and training

Increased efficiency and yields; safety

Non real-time processes inside the factory

• Tracking products andmachine inventory

• Non-real-time sensor data• Remote inspection

and diagnostics

Optimized management ofproduction facilities

Enterprise communication

• Logistics and warehousing• Employee and

back-office communications• Tracking

goods post-production

Improved business operations

3.3. Cloud Control of Machines

The capabilities afforded by 5G will enable manufacturers to design robots that benefit fromincreased data integration and real-time decision making. For a considerable length of time,manufacturing plant computerization has depended on programmable logic controllers (PLCs)that were introduced to control the machines they were connected to and afterwards hardwiredinto computing systems to implement control management. Cloud control management is mostlyvirtualized computing resources, storage, applications, and services that are managed by softwareso the data generated can be accessed on-demand. Cloud management is a combination of software,automation, policies, governance, and algorithms that determine how those cloud computing servicesare offered.

Telecom 2020, 1 55

When 5G reliably meets its initial specifications, the PLC will be able to be virtualized in thecloud, empowering machines to be controlled remotely and continuously with reduced expense to thecompany or end-user.

3.4. Expanded Reality

Factories are home to multiple advanced technological solutions designed to overcome manyproblems; one such solution is virtual reality/augmented reality. Manufacturers who have been skepticalof implementing augmented and virtual reality technologies will be able to take full advantage andleverage them for real-time simulations and predictive maintenance. 5G connectivity has the potentialto unleash massive amounts of innovation in virtual experiences that require low latency and ultra-fastdata speeds. The stability and low latency that 5G networks will bring and the use of a range offrequencies, as well as advancements within both hardware and software, is making industry veryexcited about the future use cases that will utilize this technology. One of the most pertinent issuesfacing manufacturers is the loss of skilled workers as manufacturing transitions towards digitalization.Virtual reality/augmented reality provide a way to minimize the loss of the expertise of experiencedworkers by placing them in the mentoring role of remote experts capable of guiding newer techniciansand engineers.

5G enables manufacturing organizations to extract the potential of augmented reality solutions,ultimately leading to increased productivity. Equipment maintenance is crucial to the smooth runningof most factories. The unexpected breakdown of equipment can cost companies significant amountsdepending on how long the process is halted. Engineers can use the historical data to counteract this,but this alone will still lead to excess downtime, e.g., the use of smart glasses will allow access to theconstantly generated equipment data which will allow more in-depth analysis to highlight problemequipment well before it fails. This will allow companies to choose when to maintain the equipment tominimize the financial impact. Repair manuals can be accessed in real-time, and a walkthrough of howto get access to or remove and maintain parts can be displayed to the technician or engineer. This willlead to more efficient repairs and significantly less mentoring for training staff. The best informationand work guidelines can, in this manner, be imparted to all relevant staff precisely when they need it,building specialist abilities all the more rapidly, securely, and successfully than at any other time.

4. Infrastructure Requirements for 5G

According to Singh et al. [19], attempts have been made to characterize the broad set of requirementsof industrial automation. The integration of 5G to an existing network architecture will require theaddition of an onsite 5G network comprised of macro or small cells, which will need to be fullyIP-enabled. If an existing private 4G network is in situ, the 4G backhaul network can be reusedwhenever available and possible, but backhaul capacity will likely need to be upgraded since the5G network delivers far more traffic than any cellular network did before. 5G mobile networks willsignificantly affect both the wireless side and the wired side of network infrastructure requirementsdue to the increased data traversing to and from the wired and wireless networks. The smart factoryis represented below in Figure 4 by being divided into several layers with different network choicesat each layer; each network has different demands and rates the importance of various propertiesdifferently. It is decisions at each of these levels that will dictate the level of integration that 5G can bringto manufacturing. The key new concept of network slicing in 5G will enable tenants to gain differentlevels of connectivity from their service provider to accommodate various use cases. To achieve thisnetwork slicing, 5G will be an all-cloud architecture and this will ultimately require the use of softwaredefined networking (SDN) and network function virtualization (NFV) [20].

Future 5G networks will have to provide a significantly higher system capacity than today andsolve the anticipated spectrum crunch. There will be many infrastructure improvements requiredto provide a 5G network within a manufacturing environment. Requirements will range from the

Telecom 2020, 1 56

upgrading of backbone infrastructure within the organization to managing the utilization of thefollowing:

• Spectrum adoption;• Fiber rollout internally (10 Gb minimally);• High-speed switches and routers;• On-site computing;• High-speed uplink to cloud computing facilities;• Deploying edge-connecting devices.Telecom 2020, 1,  9 

 Figure 4. Hierarchical architecture for smart factory [21]. 

A specific challenge will be the analysis of the placement of network intelligence either at the edge, in the fog, or in a private/public cloud. The precise locating of edge analytics anywhere from the network edge up to the edge cloud computing (ECC) will be a challenge for many sectors. To progress the development of the digital factory, there is a need for a robust wireless network that can transfer large amounts of data fast. Reliable low latency and high data‐transfer rates are crucial in connecting smart devices in Industry 4.0. The prospect of download speeds between 1 Gbps and 10 Gbps,  and  upload  speed  or  latency  of  just  1 millisecond, makes  the  prospect  of  a  5G  network fascinating for industry, but the reality is that typically 5G users will not achieve anything near the theoretical  top speeds. The actual rate of a 5G connection will depend on many  factors  including location, network connection type (private or via a TELephone COmpany, TELCO), how many other devices are connecting, and the specifications of the device being used. The aim for manufacturing has to be achieving a minimum download speed of 50 Mbps and latency of 10ms in a scalable manner which will represent a significant improvement over current rates. 

4.1. 5G in a Heterogeneous Network Infrastructure 

Future 5G networks will have to provide a significantly higher system capacity than today and solve  the  anticipated  spectrum  crunch.  Communications  regulators  globally  are  managing  the allocation  of  spectrum  for  5G  usage;  typically,  the  spectrum  is  being  auctioned  to  telephone companies (Telcos) who are then offering a range of solutions to  industry. In Europe, three major frequency bands have been made available for 5G use. These three bands are 700 MHz, 3.6 GHz and 26 GHz. It should be noted that spectrum regulators in some countries have started the process of opening up shared spectrum  for  local use  in specific bands, which will enable  the deployment of private 5G networks and enable easier research and development deployments. For example,  the 

Figure 4. Hierarchical architecture for smart factory [21].

A specific challenge will be the analysis of the placement of network intelligence either at theedge, in the fog, or in a private/public cloud. The precise locating of edge analytics anywhere from thenetwork edge up to the edge cloud computing (ECC) will be a challenge for many sectors. To progressthe development of the digital factory, there is a need for a robust wireless network that can transferlarge amounts of data fast. Reliable low latency and high data-transfer rates are crucial in connectingsmart devices in Industry 4.0. The prospect of download speeds between 1 Gbps and 10 Gbps,and upload speed or latency of just 1 millisecond, makes the prospect of a 5G network fascinatingfor industry, but the reality is that typically 5G users will not achieve anything near the theoreticaltop speeds. The actual rate of a 5G connection will depend on many factors including location,network connection type (private or via a TELephone Company, TELCO), how many other devicesare connecting, and the specifications of the device being used. The aim for manufacturing has to be

Telecom 2020, 1 57

achieving a minimum download speed of 50 Mbps and latency of 10 ms in a scalable manner whichwill represent a significant improvement over current rates.

4.1. 5G in a Heterogeneous Network Infrastructure

Future 5G networks will have to provide a significantly higher system capacity than todayand solve the anticipated spectrum crunch. Communications regulators globally are managing theallocation of spectrum for 5G usage; typically, the spectrum is being auctioned to telephone companies(Telcos) who are then offering a range of solutions to industry. In Europe, three major frequency bandshave been made available for 5G use. These three bands are 700 MHz, 3.6 GHz and 26 GHz. It shouldbe noted that spectrum regulators in some countries have started the process of opening up sharedspectrum for local use in specific bands, which will enable the deployment of private 5G networks andenable easier research and development deployments. For example, the regulator in the USA openedup the 3.5 GHz band as of January 2020 for commercial use known as the CBRS (Citizens BroadbandRadio Service) band.

In addition to cellular advances, Wi-Fi networking is also getting significant upgrades andthis supplementing technology will add to the 5G ecosystem. Each technology offers individualimprovements; they will also work in tandem to fill in gaps in environments where other technologiesmay not be ideal. The Wi-Fi Alliance is referring to the IEEE 802.11ax specification as Wi-Fi 6 becauseit’s the sixth generation of Wi-Fi.

Harnessing 5G’s capabilities in terms of security, synchronizing and network slicing times betweendevices will be a challenge for industry. These challenges are expected to take some time to be solved.Therefore, while 5G should be seen as the solution going forward, the timeline to address those issuesshould make companies plan a transition approach to adopting 5G, and utilising 4G LTE is a viablesolution as a transition to 5G allowing the easier and potentially more affordable adoption of 5G whenit is more mainstream.

4.2. Fog and Edge Computing

Edge computing is currently being implemented in many industries utilizing the industrybackbone for data management known as the programmable logic controller (PLC). The main taskof a PLC is to acquire data from field devices, apply logic or arithmetic functions and set outputvalues based on these computations. This typically requires round-trip times in the range of severalmicroseconds to milliseconds. With the use of 5G networks, the rate at which instructions are processedby edge networks will increase with the flow of more considerable amounts of data. Fog computingbrings intelligence to the shop floor network; edge computing brings intelligence, processing power,and communication capabilities directly into devices such as smart sensors and gateway devices [2].

4.2.1. Edge Computing

The IIoT edge gateway, or aggregation point (AP), primarily provides connectivity to networkswith a large number of sensors, controllers and actuators, all of which can use different wireless/wiredtechnologies. It is located on the edge of the network close to the data provider and the edge of thesystem leading from the machine to the cloud [2,22].

In an original conceptual IoT topology, smart sensors would collect data, and send it back to thecloud for processing. One of the critical enablers for the IoT market is the proliferation of sensors andthe actuator. These sensors need to be capable of many features whilst meeting the industry demands,which are mainly; the reduction the price of sensors, adding compute capability and connectivitycapabilities through either their wired (RS232/422/485 etc.), wireless (Wi-Fi, SigFOX, Lora, BLE, ZigBeeetc.) or radio (3G and 4G/LTE etc.) technologies. These features will considerably increase the value ofthese devices, and more importantly enable real-time management of the devices especially in largesensor arrays.

Telecom 2020, 1 58

In manufacturing environments, programmable logic circuits (PLCs) already undertake thecontext and compute capabilities, however frequently do not have connectivity capabilities or if theydo, they are proprietary in nature [23]. Conversely one of the main reasons the Internet of Things isbecoming more prevalent is due to the reduced price of commodity sensors and actuators [24] addingconnectivity and compute capabilities would provide pricing and implications for large sensor arrays.

Edge devices provide capabilities, correlating data from multiple input protocols and mediums,undertaking local processing, such as data compression, or decision making using artificial intelligencealgorithms [25], while providing connectivity to TCP/IP via IP Networks, Wi-Fi or radio accessnetworks (RAN) and interfacing to proprietary networking and industrial connectivity protocols.Another crucial role of edge compute devices, or gateways, is to provide continuous operation, runningrulesets locally even during outages or when latency issues occur on the WAN link back to the Cloudinfrastructure [26].

4.2.2. Multi-Access Edge Computing (MEC)

Multi-access edge computing (MEC), in the past known as mobile edge computing is a variantof edge computing that brings the capabilities of cloud computing to the edge of the network. Thisallows sharing of network and compute capabilities virtually anywhere, and will enable reduceddeployment time and centralized management of remote compute capabilities. Whereas traditionalcloud computing occurs typically on remote virtual servers that are situated far from the facility andthe device, MEC allows processes to take place in aggregation points on the network, bringing decisionmaking closer to the production environment. By opening this MEC environment up to more thanjust cellular users, ETSI (European Telecommunications Standards Institute) speculates that “Thecapabilities and characteristics offered by a MEC platform can be leveraged in a way that will allowproximity, agility and speed to be used for broader innovation that can be translated into unique valueand revenue generation [27]. According to research by Tran, et al. [28,29] the use of MEC and theuse of commodity server platforms to provide MEC capabilities identified five critical areas for MECarchitectures to be successful; see Table 4.

Table 4. Key areas for multi-access edge computing (MEC) architectures [2].

Deployment Scenarios and Resource Management How to identify where the best locations to install the MECplatforms physically and how to manage them remotely.

Computational Caching and Offload

The combination of computing and storage locally can providean increase in compute capability and speed to the user, whilstreducing traffic to the cloud connections. There are challengeshere around deciding what to cache, as each user’s computingrequirements may have unique data requirements.

IoT Applications and Big Data Analytics MEC can help reduce and refine datasets before sending backto big data platforms for analysis/deep learning.

Mobility Management

Ensuring continuous connectivity for mobile users is a crucialconsideration for MEC, with trajectory predictions (of the user)enabling the caching of data amongst MEC nodes in thenetwork.

Security and Privacy

As there are multiple compute points, at remote locations, theattack surface is increased exponentially over a traditionalcloud platform. Privacy of the user’s computing requirementsis also a key concern for MEC systems.

5. The Wireless Spectrum and 5G

With the rapid development of wireless communication technologies, the need for bandwidth isincreasing. Usage of the wireless spectrum has increased in recent years and that has resulted in thenecessity to manage the spectrum to allow multiple end-user devices to communicate in a structured

Telecom 2020, 1 59

method enabling interoperability across a wide array of wireless protocols both presently availableand those that are due for release in the near future [30–32].

In accordance with European law, the radio spectrum is awarded to a technology and is appliedin a technology neutral manner, this means that licensees may use the radio spectrum as they see fitprovided that they comply with the relevant applicable harmonized standards in the band in whichthey communicate. In Ireland, radio frequencies are treated as a national resource and their licensing ismanaged by the national communications Regulator, ComReg [33–37].

When an end user such as a company plans to deploy a wireless technology in a licensed band,the procedure for this is that the company will require a license from the government regulator tooperate in the chosen spectrum, in the geographical area assigned for the deployment. It is importantto note that this process only applies when operating in bands of the spectrum that are licensed.No license is required to operate in unlicensed frequency bands as long as the equipment is certifiedand approved for use in the jurisdiction. Examples of unlicensed bands would be the 418 MHz and433 Mhz channels often used for remote controls etc. [37–39].

The development of wireless communication technology has driven the need for bandwidth,and that growing need is pushing the wireless spectrum resources to become more scarce. To addressthe expected growth in traffic in the next iteration of wireless networks, there is now an increasinginterest in modifying the traditional allocation of the spectrum towards a more dynamic usage model,where the spectrum is treated as a shared resource which will allow for the deployment of multiplewireless networks in a heterogeneous architecture within a limited site of operation. The plannedheterogeneous approach is advocating a transitional approach from the static allocation of spectrum,to a more dynamic approach, which will free up the spectrum’s resources. This shared infrastructureapproach could have a transformative effect on how factories deploy their wireless assets in thefuture. Spectrum sharing in future wireless networks will likely occur in both licensed bands (e.g., the2.3–2.4 GHz band in Europe and the 3.55–3.65 GHz band in the USA) and unlicensed bands [40–43].A summary of the main wireless protocols and their attributes is in Table 5.

Table 5. Comparison of wireless protocols.

Protocol Frequency Wavelength Range

Wi-Fi (2.4G) 2401 MHz–2483 MHz 12 cm 150 m

Wi-Fi (5G) 5150 MHz–5875 MHz 5 cm 120 m

Bluetooth 2400 MHz–2485 MHz 12 cm 10–100 m

LoRA 868 MHz 35 cm 10 km+

3G 900/2100 MHz 33.3/14 cm 100 m to >5 km

4G (LTE) 800/1800 MHz 37.5/17 cm 100 m to >5 km

5G 700 MHz/3.6 GHz/26GHz 43 cm/8 cm/1 mm 50 m to >5 km

Communications regulators globally are managing the allocation of spectrum for 5G usage;typically, the spectrum is being auctioned to telcos who are then offering a range of solutions to industry.In Europe, three major frequency bands have been made available for 5G use. These three bands are700 MHz, 3.6 GHz and 26 GHz as shown in Table 5.

When considering the heterogeneous approach to wireless networks it is important to understandthat interaction of the wireless spectrum with cellular technologies will be a dominant driver for manysites. It should be noted that spectrum regulators in some countries have started the process of openingup shared spectrum for local use in specific bands, which will enable the deployment of private 5Gnetworks and enable easier research and development deployments. For example, the regulator in theUSA opened up the 3.5 GHz band as of January 2020 for commercial use known as the CBRS (CitizensBroadband Radio Service) band.

Telecom 2020, 1 60

6. Private 5G Networks

For many of the world’s businesses, private 5G will likely become the preferred choice, especially forindustrial environments such as, manufacturing plants, logistics centres, and ports. A mobile private 5Gnetwork, as the name suggests, is a local area network that utilises 5G technology as its communicationmedium to build and create a ‘private’ network.

Private 5G networks are in some circumstances referred to as local 5G networks or mobile privatenetworks (MPN), and offer unified connectivity with various advantages and optimized services.A private network that is created for an organisation is expected to carry all the features of 5Gpublic networks, including the reduced latency and higher speeds. Deploying a private networkdelivers several advantages in terms of efficiency and security this is why companies, regulators,organizations and other potential users are currently looking ahead to the arrival of private 5G networksto get the most out of this technology. Private 5G networks have the potential to become the futurecommunication platform of the factory as they will be able to provide for the increase in bandwidthrequirements coming from more connected equipment and the corresponding data being transmittedat the factory floor, as well as the millisecond latency required for real-time remote control of robotics.

Since private 4G LTE networks are already commercial and in extensive use, deploying private5G networks is obviously an expectation that will allow manufacturers to eliminate the tetheredlimitations of ethernet cables that litter their factory floors and connect the machines wirelessly to thecloud. The capacity of a network due to the number of IoT and wireless devices to be connected on anetwork will be considerably higher once a 5G network is deployed due to the increased bandwidth of5G [44]. Due to the wide variations of IoT requirements, several overlapping topologies exist, from thesensor or actuator through local environments, to decentralized or core cloud capabilities, and reachingout to public, private or hybrid clouds [45]. A significant challenge for 5G in business terms is ifthe plant owners and operators are willing to pay for a licensed spectrum instead of using availableunlicensed spectrum technology, which is effectively a potential cost-free improvement. To enableenterprise connectivity with ultra-reliable, high-speed, low-latency, power-efficient, high-densitywireless connectivity a company essentially has two options:

• Connect to a public 5G network.• Build and use a private 5G network.

Deployments in industry have been mainly in the two formats mentioned above. These involvethe deployment of a physically-isolated private 5G network (a 5G island) that is independent of themobile operator’s public 5G network. Audi, for example, has already started testing 5G for roboticmotion control and Nokia is using 5G for in-factory private 5G connectivity. Another example ofthe private network deployment model is the e.GO car company in Aachen, Germany. In e.GO’sfactory, where the e.GO Life model is manufactured, an Ericsson private network with a 5G core and5G new radio solution has been deployed to deliver secure and near real-time data networking acrossthe production chain, from digital material management to autonomous vehicle control. The on-siteprivate 5G network has helped establish an automated factory realizing and delivering connectivityfor an array of connected devices such as machines, sensors, materials and robots can be managedthrough one standardized network, with the correct latency and bandwidth allocated as necessary.

As 5G has developed it has been adopted by companies eager to test and potentially take firstmover advantage. The second deployment scenario being adopted by companies is a more basicapproach, to connect to a public 5G network, sharing the mobile operator’s public 5G network withother users. An example of this scenario is when companies are connecting to the 5G network dueto inadequate Wi-Fi connectivity options. In both deployment models; for 5G to be successful in amanufacturing environment there must be a close collaboration between all systems, spanning acrossthe corporate IT and the manufacturing OT to maximize the capabilities inherent in 5G [46–49].

For deployment of a private 5G network it would involve a company purchasing an infrastructurewhile contracting for operational support usually from a mobile operator. Or the company can build

Telecom 2020, 1 61

and maintain their own 5G network using their own specific spectrum, obtained from the spectrumregulator in the country where the 5G network will operate. An example of a proposed layout for4G/5G deployment in a corporate network is shown in Figure 5.Telecom 2020, 1,  14 

 Figure 5. Private 5G infrastructure supporting Industry 4.0. 

Benefits of Private 5G Networks 

Non‐public 5G networks turn businesses into mobile operators in limited geographic areas. A company that operates a mobile private 5G network (MPN) will become a 5G network infrastructure operator; this will probably be the case if the organisation wants to have complete control over its operating processes. As a result, parameters such as the bandwidth or real‐time properties can be adapted, which is not possible when using the public network of a mobile radio provider.   

An organisation that runs a private 5G network will be able to add  layered security at every level of the manufacturing process. Since the security can be higher, it will afford the owner of the network a level of control that would not be possible on a public network. These networks also offer protection  against  industrial  espionage. Data  in  non‐public  5G  networks  can  be  segregated  and processed  separately  from public 5G networks which will ensure  complete privacy protection of process  and  production‐related  data  because  the  communication  takes  place within  a  separate network, enabling the data to be far better protected against unauthorised access [50,51]. 

Unlike a public network, a private 5G network  can be  configured  to  the  specific needs of a location, and configurations can vary by site, depending on  the  type of work undertaken  in each venue. A private network allows companies to determine the network’s deployment timetable and coverage quality. Another point of note is that the network may be installed and maintained by onsite personnel, enabling faster responses to issues, thereby reducing downtime and maintaining a level of uptime to meet the organisation’s needs. 

7. Security in the Wireless Factory of the Future 

One of the initial concerns that companies have about deploying a 5G network to a site that will have connectivity to thousands of devices is security, because 5G connected devices will dramatically expand the potential network‐attack surface, Figure 6. A smart factory’s system will include countless pieces  of  equipment  and  devices  that  are  connected  to  a  single  organisational  network. Vulnerabilities in any one of those devices can create a gap in the network that could open the system up to security threats. From a factory viewpoint, OT systems have predominantly been isolated from external networks for many decades. However, due to the introduction of TCP/IP and other network communication  standards within OT environments,  the OT elements have become  susceptible  to cyber‐attacks in recent years [52,53].   

Figure 5. Private 5G infrastructure supporting Industry 4.0.

Benefits of Private 5G Networks

Non-public 5G networks turn businesses into mobile operators in limited geographic areas.A company that operates a mobile private 5G network (MPN) will become a 5G network infrastructureoperator; this will probably be the case if the organisation wants to have complete control over itsoperating processes. As a result, parameters such as the bandwidth or real-time properties can beadapted, which is not possible when using the public network of a mobile radio provider.

An organisation that runs a private 5G network will be able to add layered security at everylevel of the manufacturing process. Since the security can be higher, it will afford the owner of thenetwork a level of control that would not be possible on a public network. These networks alsooffer protection against industrial espionage. Data in non-public 5G networks can be segregatedand processed separately from public 5G networks which will ensure complete privacy protection ofprocess and production-related data because the communication takes place within a separate network,enabling the data to be far better protected against unauthorised access [50,51].

Unlike a public network, a private 5G network can be configured to the specific needs of alocation, and configurations can vary by site, depending on the type of work undertaken in each venue.A private network allows companies to determine the network’s deployment timetable and coveragequality. Another point of note is that the network may be installed and maintained by onsite personnel,enabling faster responses to issues, thereby reducing downtime and maintaining a level of uptime tomeet the organisation’s needs.

7. Security in the Wireless Factory of the Future

One of the initial concerns that companies have about deploying a 5G network to a site that willhave connectivity to thousands of devices is security, because 5G connected devices will dramaticallyexpand the potential network-attack surface, Figure 6. A smart factory’s system will include countlesspieces of equipment and devices that are connected to a single organisational network. Vulnerabilities inany one of those devices can create a gap in the network that could open the system up to security threats.From a factory viewpoint, OT systems have predominantly been isolated from external networks

Telecom 2020, 1 62

for many decades. However, due to the introduction of TCP/IP and other network communicationstandards within OT environments, the OT elements have become susceptible to cyber-attacks in recentyears [52,53].

As factories of the future transition increasingly towards wireless infrastructures, both currentlyavailable and due for release, they will have to build a comprehensive cybersecurity plan that addressesthe threats resulting from IT/OT convergence, such as:

• Denial-of-service (DoS) and distributed denial-of-service (DDoS) attacks on production systems.• Man-in-the-middle attack for production data theft due to gaps in the IT structure of the newly

adapted manufacturing processes.• Anomalies based on human behaviour and errors.• Critical system access to third-party contractors.• Legacy devices with limited or no security services such as programmable logic controllers

(PLCs), remote terminal units (RTUs), supervisory control and data acquisition (SCADA) servers,integrated with fully accessible and network-enabled IoT devices.

• Lack of adequate visibility and cyber-resilient systems. Traditionally OT systems have achievedsecurity by obscurity and this is no longer a guarantee of protection with the advent of wirelessconnectivity and massive machine-type communication (mMTC).

Telecom 2020, 1,  15 

As factories of the future transition increasingly towards wireless infrastructures, both currently available  and  due  for  release,  they will  have  to  build  a  comprehensive  cybersecurity  plan  that addresses the threats resulting from IT/OT convergence, such as: 

Denial‐of‐service (DoS) and distributed denial‐of‐service (DDoS) attacks on production systems.    Man‐in‐the‐middle attack for production data theft due to gaps in the IT structure of the newly 

adapted manufacturing processes.    Anomalies based on human behaviour and errors.    Critical system access to third‐party contractors.    Legacy devices with  limited or no  security  services  such  as programmable  logic  controllers 

(PLCs),  remote  terminal  units  (RTUs),  supervisory  control  and  data  acquisition  (SCADA) servers, integrated with fully accessible and network‐enabled IoT devices.   

Lack of adequate visibility and cyber‐resilient systems. Traditionally OT systems have achieved security by obscurity and this is no longer a guarantee of protection with the advent of wireless connectivity and massive machine‐type communication (mMTC). 

 Figure 6. The 5G threat landscape. 

Historically, successful transitions into new technologies have always been focused on end goal results with  security  rarely at  the  forefront. While 5G has  incorporated  security  into  its design a logical  approach  for  smart manufacturing  should  be  to develop  a more  robust  suite  of  security protocols at the outset [54]. This will allow for an easier adoption by manufacturers and for a more robust end user  system  that would allow  for multiple wireless  technologies operating  in parallel across  a production  floor. As giving  a  level of  trustworthiness  in  the  system  to meet  regulatory requirements will be an issue when systems are deployed in regulated manufacturing environments, a  model  that  demonstrates  trustworthiness  and  reliability  will  need  to  be  honed  for  the manufacturing sector. 

8. Standards and Industry Associations 

The 5G wireless standard aims to be global which is difficult due to different implementation approaches in multiple countries, e.g., Russia, China and South Korea or an amalgamated body of countries (e.g., the European Union (EU)) outlining their own definition of 5G networks, speeds and regulations  outlining  where  5G  transmissions  may  occur.  In  broad  terms,  standards  benefit innovation, the environment, safety and the economy. Many standards bodies are contributing to the 

Figure 6. The 5G threat landscape.

Historically, successful transitions into new technologies have always been focused on end goalresults with security rarely at the forefront. While 5G has incorporated security into its design a logicalapproach for smart manufacturing should be to develop a more robust suite of security protocolsat the outset [54]. This will allow for an easier adoption by manufacturers and for a more robustend user system that would allow for multiple wireless technologies operating in parallel across aproduction floor. As giving a level of trustworthiness in the system to meet regulatory requirementswill be an issue when systems are deployed in regulated manufacturing environments, a model thatdemonstrates trustworthiness and reliability will need to be honed for the manufacturing sector.

Telecom 2020, 1 63

8. Standards and Industry Associations

The 5G wireless standard aims to be global which is difficult due to different implementationapproaches in multiple countries, e.g., Russia, China and South Korea or an amalgamated body ofcountries (e.g., the European Union (EU)) outlining their own definition of 5G networks, speeds andregulations outlining where 5G transmissions may occur. In broad terms, standards benefit innovation,the environment, safety and the economy. Many standards bodies are contributing to the developmentof 5G for manufacturing, e.g., The 5G Alliance for Connected Industries and Automation (5G-ACIA) isa global forum for collaboration. A summary list is given below.

• Industrial Internet Consortium (IIC), https://www.iiconsortium.org/index.htm.• The 3rd Generation Partnership Project (3GPP), https://www.3gpp.org/.• The 5G Alliance for Connected Industries and Automation (5G-ACIA), https://www.5gacia.org/

about-5g-acia/.• EU 5GPPP, https://5g-ppp.eu/.

Industry, whilst working in conjunction with standards, has many open questions still toinvestigate, such as:

• Co-existence of different wireless protocols and systems;• Co-existence of different wired protocols;• Interoperability between wired and wireless communication systems;• Seamless engineering taking into account collected real-life data;• Missing capabilities for plug-and-produce integration of sensors, machinery and people;• Missing capabilities for cost-effective network and service management by factory operator;• Managing workflows and data interaction patterns between an increasing number of sensors,

machines, robots, wearables, etc.;• Allocating the proper computing resources in the cloud, on the edge or at sensor levels to ensure

compliance with application-level service agreements;• Leveraging machine-learning and data analytics capabilities distributed in the network,

to a multitude of vendor-specific platforms, contributing to a unified, intelligent dataintelligence platform;

• Managing and optimizing the wireless network topology and performance, according to real-timechanging networking conditions.

Future Improvements of 5G towards 6G

While 5G will be improved upon, 6G is expected to be launched commercially in 2030. It isa technology that is being developed in response to the increasing availability of distributed radioaccess network (RAN) and the desire to take advantage of the terahertz spectrum (THz) to furtherincrease capacity and reduce latency. It is expected that many of the perceived problems associatedwith deploying millimeter-wave radio for 5G radio will be resolved in time for network designers toaddress the challenge of implementing a 6G offering. 6G connectivity is expected to support speedsof 1 terabyte per second (Tbps). This level of capacity and latency will be unprecedented and willextend the performance of 5G applications along with expanding the scope of capabilities in supportof increasingly new and innovative tools in the areas of wireless cognition, detection and imaging.In addition, 6G wireless sensing solutions are expected to selectively use different frequencies tomeasure absorption and adjust frequencies accordingly in order to maximize signal penetration andtransmission distances.

9. Conclusions

All of the technological and business monitors are indicating that 5G has the potential torevolutionize manufacturing and, over time, will improve productivity. While 5G is capable of bringing

https://www.iiconsortium.org/index.htmhttps://www.3gpp.org/https://www.5gacia.org/about-5g-acia/https://www.5gacia.org/about-5g-acia/https://5g-ppp.eu/

Telecom 2020, 1 64

many solutions, there is also a myriad of engineering matters that need to be considered before 5Gcan be the all-conquering panacea for manufacturing. While there are many barriers to entry forindustry adopting the suite of technologies incorporated in 5G, ranging from, spectrum allocation,scalability issues, edge analytics and security concerns which have been detailed in this paper,further investigation is required into the adoption of an unbounded wireless transmission medium in asociety that is increasingly conscious of data privacy and that will be a challenge not just for engineersbut also standards bodies and lawmakers.

When considering a 5G deployment, industry needs to assess the specific requirements for eachIoT use case. Industry needs to get a clearer understanding of the data types being transmitted,their sensitivity to latency, and whether the data needs to be processed by itself, or against historicalinformation. The answers to these requirements and other factors that will dictate whether or not acompany needs to implement a smart manufacturing solution using 5G topologies or whether thereare other protocols that may be more suited to the individual solution.

5G will not only advance manufacturing operations by enabling businesses to be faster but willalso make them more responsive to customer needs. Manufacturers will benefit from the deploymentof 5G since warehouses, factories and other facilities will increasingly have been enabled with smarttechnologies that will benefit from higher speed networks. 5G’s unprecedented speed and coveragewill bring the world closer than ever, as well as providing an avenue for never before imaginedcapabilities across the manufacturing space. It is an exciting time for manufacturing and for anycompany that can leverage the improvements that 5G and beyond offers.

Finally, it is clear that further research should be undertaken into the interoperability of wirelesstechnologies in factory environments to reveal the reliability, latency, and minimum range of wirelesstechnologies in a factory environment when deployed in conjunction with 5G, evaluating theheterogeneous interoperability of 5G and existing communication protocols currently deployedin manufacturing environments.

Funding: The authors would also like to acknowledge resources made available through the Science FoundationIreland through the grant award (16/RC/3918) to the Confirm Centre for Smart Manufacturing.

Acknowledgments: The authors would like to thank the staff of the Electronic and Computer EngineeringDepartment at the University of Limerick for their assistance in this work.

Conflicts of Interest: The authors declare no conflict of interest.

References

1. Mao, Y.; You, C.; Zhang, J.; Huang, K.; Letaief, K.B. A Survey on Mobile Edge Computing: The CommunicationPerspective. IEEE Commun. Surv. Tutor. 2017, 19, 2322–2358. [CrossRef]

2. McCann, J.; Quinn, L.; McGrath, S.; O’Connell, E. Towards the Distributed Edge–An IoT Review.In Proceedings of the 2018 12th International Conference on Sensing Technology (ICST), Limerick, Ireland,3–6 December 2018; pp. 263–268.

3. Bockelmann, C.; Pratas, N.K.; Wunder, G.; Saur, S.; Navarro, M.; Gregoratti, D.; Vivier, G.; De Carvalho, E.;Ji, Y.; Stefanovic, C.; et al. Towards Massive Connectivity Support for Scalable mMTC Communications in5G Networks. IEEE Access 2018, 6, 28969–28992. [CrossRef]

4. Popovski, P.; Trillingsgaard, K.F.; Simeone, O.; Durisi, G. 5G Wireless Network Slicing for eMBB, URLLC,and mMTC: A Communication-Theoretic View. IEEE Access 2018, 6, 55765–55779. [CrossRef]

5. Siddiqi, M.; Yu, H.; Joung, J. 5G Ultra-Reliable Low-Latency Communication Implementation Challengesand Operational Issues with IoT Devices. Electronics 2019, 8, 981. [CrossRef]

6. Gidlund, M.; Lennvall, T.; Akerberg, J. Will 5G become yet another wireless technology for industrialautomation? In Proceedings of the 2017 IEEE International Conference on Industrial Technology (ICIT),Toronto, ON, Canada, 22–25 March 2017; pp. 1319–1324.

7. Orlosky, J.; Kiyokawa, K.; Takemura, H. Virtual and Augmented Reality on the 5G Highway. J. Inf. Process.2017, 25, 133–141. [CrossRef]

http://dx.doi.org/10.1109/COMST.2017.2745201http://dx.doi.org/10.1109/ACCESS.2018.2837382http://dx.doi.org/10.1109/ACCESS.2018.2872781http://dx.doi.org/10.3390/electronics8090981http://dx.doi.org/10.2197/ipsjjip.25.133

Telecom 2020, 1 65

8. GSMA Intelligence. Understanding 5G: Perspectives on Future Technological Advancements in Mobile.White Paper. 2014, pp. 1–26. Available online: https://www.gsma.com/futurenetworks/wp-content/uploads/2015/01/2014-12-08-c88a32b3c59a11944a9c4e544fee7770.pdf (accessed on 10 June 2020).

9. Cheng, J.; Chen, W.; Tao, F.; Lin, C.-L.; Da Xu, L. Industrial IoT in 5G environment towards smartmanufacturing. J. Ind. Inf. Integr. 2018, 10, 10–19. [CrossRef]

10. Mudrakola, S. Private 5G Networks. 2019. Available online: http://techgenix.com/private-5g-networks/(accessed on 10 June 2020).

11. Onoe, S. Plenary speaker 1. In Proceedings of the 2017 IEEE Radio Frequency Integrated Circuits Symposium(RFIC), Honolulu, HI, USA, 4–6 June 2017; p. 1. [CrossRef]

12. Atanasovski, V.; Leon-Garcia, A. Future Access Enablers for Ubiquitous and Intelligent Infrastructures; Springer:Berlin, Germany, 2015.

13. Baharom, B.; Ali, M.T. Multiple-element PIFA MIMO antenna system design for future 5G wirelesscommunication applications. In Proceedings of the 2017 IEEE Asia Pacific Microwave Conference (APMC),Kuala Lumpur, Malaysia, 13–16 November 2017; pp. 743–746.

14. Arthur D. Little for Vodafone Group Plc. “Creating a Gigabit Society”. Availableonline: https://www.vodafone.com/content/dam/vodcom/files/public-policy/Vodafone_Group_Call_for_the_Gigabit_SocietyFV.pdf (accessed on 10 June 2020).

15. Rampa, V.; Savazzi, S.; Malandrino, F. Opportunistic sensing in beyond-5G networks: The opportunities oftransformative computing. In 5G Italy Book 2019: A Multiperspective View of 5G; CNIT: Parma, Italy, 2019;pp. 461–476. ISBN 9788832170030.

16. Wdowik, R.; Ratnayake, R.C. Collaborative Technological Process Planning with 5G Mobile Networks andDigital Tools: Manufacturing Environments’ Perspective. In Proceedings of the 2019 IEEE InternationalConference on Industrial Engineering and Engineering Management (IEEM), Macau, China, 15–18 December2019; pp. 349–353.

17. Plant Automation Technology Impact of 5G in Manufacturing Industry. Available online: https://www.plantautomation-technology.com/articles/impact-of-5g-in-manufacturing-industry (accessed on10 June 2020).

18. Elayoubi, S.E.; Fallgren, M.; Spapis, P.; Zimmermann, G.; Martín-Sacristán, D.; Yang, C.; Jeux, S.; Agyapong, P.;Campoy, L.; Qi, Y.; et al. 5G service requirements and operational use cases: Analysis and METIS II vision.In Proceedings of the 2016 European Conference on Networks and Communications (EuCNC), Athens,Greece, 27–30 June 2016; pp. 158–162.

19. Singh, S.; Singh, P. Key concepts and network architecture for 5G mobile technology. Int. J. Sci. Res. Eng.Technol. (IJSRET) 2012, 1, 165–170.

20. Barakabitze, A.A.; Ahmad, A.; Mijumbi, R.; Hines, A. 5G network slicing using SDN and NFV: A survey oftaxonomy, architectures and future challenges. Comput. Netw. 2020, 167, 106984. [CrossRef]

21. Chen, B.; Wan, J.; Shu, L.; Li, P.; Mukherjee, M.; Yin, B. Smart Factory of Industry 4.0: Key Technologies,Application Case, and Challenges. IEEE Access 2018, 6, 6505–6519. [CrossRef]

22. Ludwig, S.; Karrenbauer, M.; Fellan, A.; Schotten, H.; Buhr, H.; Seetaraman, S.; Niebert, N.; Bernardy, A.;Seelmann, V.; Stich, V.; et al. A5G architecture for the factory of the future. In Proceedings of the 2018IEEE 23rd international conference on emerging technologies and factory automation (ETFA), Torino, Italy,4–7 September 2018; Volume 1, pp. 1409–1416.

23. Shi, W.; Cao, J.; Zhang, Q.; Li, Y.; Xu, L. Edge Computing: Vision and Challenges. IEEE Internet Things J.2016, 3, 637–646. [CrossRef]

24. Gubbi, J.; Buyya, R.; Marusic, S.; Palaniswami, M. Internet of Things (IoT): A vision, architectural elements,and future directions. Futur. Gener. Comput. Syst. 2013, 29, 1645–1660. [CrossRef]

25. Vermesan, O.; Bröring, A.; Tragos, E.; Serrano, M.; Bacciu, D.; Chessa, S.; Gallicchio, C.; Micheli, A.;Dragone, M.; Saffiotti, A.; et al. Internet of robotic things: Converging sensing/actuating, hypoconnectivity,artificial intelligence and IoT Platforms. In Cognitive Hyperconnected Digital Transformation: Internet of ThingsIntelligence Evolution; River Publishers: Gistrup, Denmark, 2017; pp. 97–155.

26. Gremban, K. Editorial and Introduction to the Issue: Risk and Rewards of the Internet of Things. IEEE InternetThings Mag. 2018, 1, 2–3. [CrossRef]

27. Hu, Y.C.; Patel, M.; Sabella, D.; Sprecher, N.; Young, V. Mobile edge computing—A key technology towards5G. ETSI White Paper 2015, 11, 1–16.

https://www.gsma.com/futurenetworks/wp-content/uploads/2015/01/2014-12-08-c88a32b3c59a11944a9c4e544fee7770.pdfhttps://www.gsma.com/futurenetworks/wp-content/uploads/2015/01/2014-12-08-c88a32b3c59a11944a9c4e544fee7770.pdfhttp://dx.doi.org/10.1016/j.jii.2018.04.001http://techgenix.com/private-5g-networks/http://dx.doi.org/10.1109/RFIC.2017.7969000https://www.vodafone.com/content/dam/vodcom/files/public-policy/Vodafone_Group_Call_for_the_Gigabit_SocietyFV.pdfhttps://www.vodafone.com/content/dam/vodcom/files/public-policy/Vodafone_Group_Call_for_the_Gigabit_SocietyFV.pdfhttps://www.plantautomation-technology.com/articles/impact-of-5g-in-manufacturing-industryhttps://www.plantautomation-technology.com/articles/impact-of-5g-in-manufacturing-industryhttp://dx.doi.org/10.1016/j.comnet.2019.106984http://dx.doi.org/10.1109/ACCESS.2017.2783682http://dx.doi.org/10.1109/JIOT.2016.2579198http://dx.doi.org/10.1016/j.future.2013.01.010http://dx.doi.org/10.1109/MIOT.2018.8552481

Telecom 2020, 1 66

28. Tran, T.X.; Hajisami, A.; Pandey, P.; Pompili, D. Collaborative Mobile Edge Computing in 5G Networks:New Paradigms, Scenarios, and Challenges. IEEE Commun. Mag. 2017, 55, 54–61. [CrossRef]

29. Tran, T.X.; Hosseini, M.-P.; Pompili, D. Mobile edge computing: Recent efforts and five key research directions.IEEE COMSOC MMTC Commun-Frontiers 2017, 12, 29–34.

30. Wu, Q.; Ding, G.; Du, Z.; Sun, Y.; Jo, M.; Vasilakos, A.V. A Cloud-Based Architecture for the Internet ofSpectrum Devices Over Future Wireless Networks. IEEE Access 2016, 4, 2854–2862. [CrossRef]

31. Wu, C.; Wang, Y.; Yin, Z. Energy-efficiency opportunistic spectrum allocation in cognitive wireless sensornetwork. EURASIP J. Wirel. Commun. Netw. 2018, 2018, 1–14. [CrossRef]

32. Wieruch, D.; Holfeld, B.; Wirth, T. Wireless Factory Automation: Radio Channel Evolution in RepeatedManufacturing Processes. In Proceedings of the WSA 2016, 20th International ITG Workshop on SmartAntennas, VDE, Munich, Germany, 9–11 March 2016; pp. 1–4.

33. Montgomery, K.; Candell, R.; Liu, Y.; Hany, M. Wireless User Requirements for the Factory Workcell.Available online: https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=928517 (accessed on 10 June 2020).

34. Mochón, M.A.; Saez, Y. A review of radio spectrum combinatorial clock auctions. Telecommun. Policy 2017,41, 303–324. [CrossRef]

35. Kalliovaara, J.; Ekman, R.; Paavola, J.; Jokela, T.; Hallio, J.; Auranen, J.; Talmola, P.; Kokkinen, H.Designing a testbed infrastructure for experimental validation and trialing of 5G vertical applications.In Proceedings of the International Conference on Cognitive Radio Oriented Wireless Networks, Lisbon,Portugal, 20–21 September 2017; pp. 247–263. [CrossRef]

36. Hu, F.; Chen, B.; Zhu, K. Full Spectrum Sharing in Cognitive Radio Networks Toward 5G: A Survey.IEEE Access 2018, 6, 15754–15776. [CrossRef]

37. Ejaz, W.; Ibnkahla, M. Multiband Spectrum Sensing and Resource Allocation for IoT in Cognitive 5GNetworks. IEEE Internet Things J. 2017, 5, 150–163. [CrossRef]

38. Doyle, L.; Kibilda, J.; Forde, T.K.; DaSilva, L. Spectrum Without Bounds, Networks Without Borders.Proc. IEEE 2014, 102, 351–365. [CrossRef]

39. Dong, X.; Cheng, L.; Zheng, G.; Wang, T. Network access and spectrum allocation in next-generationmulti-heterogeneous networks. Int. J. Distrib. Sens. Netw. 2019, 15, 1550147719866140. [CrossRef]

40. IEEE. 802.11-2016-IEEE Standard for Information Technology–Telecommunications and Information Exchangebetween Systems Local and Metropolitan Area Networks–Specific Requirements-Part 11: Wireless LAN MediumAccess Control (MAC) and Physical Layer (PHY) Specifications; IEEE: Piscataway, NJ, USA, 2016.

41. Bhattarai, S.; Park, J.-M.; Gao, B.; Bian, K.; Lehr, W. An Overview of Dynamic Spectrum Sharing: OngoingInitiatives, Challenges, and a Roadmap for Future Research. IEEE Trans. Cogn. Commun. Netw. 2016, 2,110–128. [CrossRef]

42. Ashraf, S.A.; Aktas, I.; Eriksson, E.; Helmersson, K.W.; Ansari, J. Ultra-reliable and low-latency communicationfor wireless factory automation: From LTE to 5G. In Proceedings of the 2016 IEEE 21st International Conferenceon Emerging Technologies and Factory Automation (ETFA), Berlin, Germany, 6–9 September 2016; pp. 1–8.[CrossRef]

43. Akyildiz, I.F.; Lee, W.-Y.; Vuran, M.C.; Mohanty, S. NeXt generation/dynamic spectrum access/cognitiveradio wireless networks: A survey. Comput. Netw. 2006, 50, 2127–2159. [CrossRef]

44. Mahmud, R.; Kotagiri, R.; Buyya, R. Fog Computing: A Taxonomy, Survey and Future Directions. In Internetof Everything; Springer: Berlin, Germany, 2018; pp. 103–130.

45. Li, J.; Li, Y.K.; Chen, X.; Lee, P.P.; Lou, W. A Hybrid Cloud Approach for Secure Authorized Deduplication.IEEE Trans. Parallel Distrib. Syst. 2014, 26, 1206–1216. [CrossRef]

46. Chien, H.-T.; Lin, Y.-D.; Lai, C.-L.; Wang, C.-T. End-to-End Slicing as a Service with Computing andCommunication Resource Allocation for Multi-Tenant 5G Systems. IEEE Wirel. Commun. 2019, 26, 104–112.[CrossRef]

47. Coll-Perales, B.; Gozalvez, J.; Maestre, J.L. 5G and Beyond: Smart Devices as Part of the Network Fabric.IEEE Netw. 2019, 33, 170–177. [CrossRef]

48. Gringoli, F.; Patras, P.; Donato, C.; Serrano, P.; Grunenberger, Y. Performance Assessment of Open SoftwarePlatforms for 5G Prototyping. IEEE Wirel. Commun. 2018, 25, 10–15. [CrossRef]

49. Di Mauro, M.; Liotta, A. An experimental evaluation and characterization of VoIP over an LTE-A network.IEEE Trans. Netw. Serv. Manag. 2020. [CrossRef]

http://dx.doi.org/10.1109/MCOM.2017.1600863http://dx.doi.org/10.1109/ACCESS.2016.2576286http://dx.doi.org/10.1186/s13638-017-1018-9https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=928517http://dx.doi.org/10.1016/j.telpol.2016.12.003http://dx.doi.org/10.1007/978-3-319-76207-4_21http://dx.doi.org/10.1109/ACCESS.2018.2802450http://dx.doi.org/10.1109/JIOT.2017.2775959http://dx.doi.org/10.1109/JPROC.2014.2302743http://dx.doi.org/10.1177/1550147719866140http://dx.doi.org/10.1109/TCCN.2016.2592921http://dx.doi.org/10.1109/ETFA.2016.7733543http://dx.doi.org/10.1016/j.comnet.2006.05.001http://dx.doi.org/10.1109/TPDS.2014.2318320http://dx.doi.org/10.1109/MWC.2019.1800466http://dx.doi.org/10.1109/MNET.2019.1800136http://dx.doi.org/10.1109/MWC.2018.1800049http://dx.doi.org/10.1109/TNSM.2020.2995505

Telecom 2020, 1 67

50. O’Connell, E.; Healy, M.; O’Keeffe, S.; Newe, T.; Lewis, E. Development of a discriminating fibre opticsensing array for wireless real time analysis of the maritime environment. In Proceedings of the OFS201222nd International Conference on Optical Fiber Sensors, Beijing, China, 14–19 October 2012. [CrossRef]

51. Ruili, J.; Haocong, W.; Han, W.; O’Connell, E.; McGrath, S. Smart Parking System Using Image Processingand Artificial Intelligence. In Proceedings of the 2018 12th International Conference on Sensing Technology(ICST), Limerick, Ireland, 3–6 December 2018; pp. 232–235.

52. Haerick, W.; Gupta, M. 5G and the Factories of the Future. 5G-PPP White Paper. 2015. Availableonline: https://5g-ppp.eu/wp-content/uploads/2014/02/5G-PPP-White-Paper-on-Factories-of-the-Future-Vertical-Sector.pdf (accessed on 10 June 2020).

53. ENISA draws Threat Landscape of 5G Networks. Available online: https://www.enisa.europa.eu/news/enisa-news/enisa-draws-threat-landscape-of-5g-networks. (accessed on 10 June 2020).

54. Vidal, J.M.; Monge, M.A.S. Framework for Anticipatory Self-Protective 5G Environments. In Proceedings ofthe 14th International Conference on Availability, Reliability and Security, Canterbury, UK, 26–29 August 2019;pp. 1–9.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open accessarticle distributed under the terms and conditions of the Creative Commons Attribution(CC BY) license (http://creativecommons.org/licenses/by/4.0/).

http://dx.doi.org/10.1117/12.975162https://5g-ppp.eu/wp-content/uploads/2014/02/5G-PPP-White-Paper-on-Factories-of-the-Future-Vertical-Sector.pdfhttps://5g-ppp.eu/wp-content/uploads/2014/02/5G-PPP-White-Paper-on-Factories-of-the-Future-Vertical-Sector.pdfhttps://www.enisa.europa.eu/news/enisa-news/enisa-draws-threat-landscape-of-5g-networks.https://www.enisa.europa.eu/news/enisa-news/enisa-draws-threat-landscape-of-5g-networks.http://creativecommons.org/http://creativecommons.org/licenses/by/4.0/.

Introduction Enhanced Mobile Broadband (eMBB) Ultra-Reliable and Low-Latency Communications (URLLC) Massive Machine-Type Communications (mMTC)

The Potential of 5G in Manufacturing Use Cases of 5G in Manufacturing Supply Chain Unification Artificial Intelligence